Evolutionary processes are fundamentally shaped by the physico-chemical properties of proteins, DNA, and other biomolecules: many layers of biophysical and biochemical mechanisms connect a DNA mutation to a cell's ability to survive and reproduce. Little is known about how random mutations affect these molecular properties and ultimately shape the evolutionary fate of a population. This knowledge is particularly crucial for predicting or controlling the evolution of antibiotic resistance in bacteria. The complexity of binding interactions and gene regulation suggests that a mutation to a single protein may have far-reaching effects throughout the cell. The goal of this project is to determine how mutations affect these molecular interactions and the resulting consequences for evolution in bacteria. Using a combination of theoretical and experimental approaches, it tests the hypothesis that protein-protein and regulatory interactions evolve rapidly and predictably in response to a mutational perturbation, while specific compensatory mutations that target individual protein traits, such as folding stability, and evolve over longer times. The project will first develop a multi-scale model that combines the biochemical kinetics of protein folding, binding, and regulation with the evolutionary dynamics of selection, mutation, and genetic drift. Computational simulations of the model will make experimental predictions. The project will then test these predictions by experimentally evolving E. coli with mutant dihydrofolate reductase (DHFR), an essential metabolic enzyme. Whole-genome sequencing and phenotyping of the evolved populations will allow comparison of the experiments with the theoretical predictions. This project will substantially increase our understanding of how mutations affect molecular interactions in cells, and how this shapes evolutionary outcomes.